13 January 2018, 14:58 | #1541 |
Registered User
Join Date: Dec 2015
Location: USA
Posts: 2,933
|
Had to call it something, and now it's going to be a pattern.
1) Rapture 2) Ragnarok 3) Dagor Dagorath Still trying to figure out the memory timings, slight improvement, hit 59.1 FPS in Quake2@640x480. Last edited by grelbfarlk; 13 January 2018 at 20:00. |
13 January 2018, 22:41 | #1542 |
Registered User
Join Date: Dec 2015
Location: USA
Posts: 2,933
|
Tried compiling flops with altivec turned on, don't know whether this is making a difference or not. Compiler options used were -warpup -O2 -mmultiple -mcpu=7400 -mabi=altivec -faltivec
Sonnet results: 8.Workbench:> flops FLOPS C Program (Double Precision), V2.0 18 Dec 1992 Module Error RunTime MFLOPS (usec) 1 2.8422e-14 0.1300 107.6830 2 2.5047e-13 0.1050 66.6545 3 -7.6605e-15 0.0925 183.7690 4 2.2782e-13 0.1200 124.9897 5 3.8969e-14 0.2100 138.0771 6 7.6050e-15 0.1325 218.8426 7 -1.1369e-13 0.3275 36.6365 8 1.2612e-13 0.1425 210.4986 Iterations = 128000000 NullTime (usec) = 0.0012 MFLOPS(1) = 84.1964 MFLOPS(2) = 76.1813 MFLOPS(3) = 126.3924 MFLOPS(4) = 186.6470 Rapture results: 7.Xworkbench:> flops FLOPS C Program (Double Precision), V2.0 18 Dec 1992 Module Error RunTime MFLOPS (usec) 1 -7.6739e-13 0.1067 131.2303 2 -5.7021e-13 0.0889 78.7361 3 -2.4314e-14 0.0622 273.1813 4 6.8612e-14 0.0889 168.7265 5 -1.6209e-14 0.1667 173.9709 6 1.3961e-13 0.0956 303.4453 7 -3.6152e-11 0.2845 42.1805 8 8.9373e-15 0.1022 293.4363 Iterations = 256000000 NullTime (usec) = 0.0004 MFLOPS(1) = 102.6142 MFLOPS(2) = 91.6743 MFLOPS(3) = 161.0047 MFLOPS(4) = 260.7921 If anyone wants to try this on some other WarpOS platform here is the executable. I don't testify that these results are indicative of anything other than running the same executable on the machines above. Just ran AFlopsWOS on both and it is faster on both machines, so Altivec compiler flags aren't doing anything. AflopsWos on Sonnet: 9.Ram Disk:> aflopswos Amiga Flops Benchmark (Double Precision), V2.0 Module Error RunTime MFLOPS (usec) 1 2.8422e-14 0.1250 112.0000 2 2.5047e-13 0.0975 71.7949 3 -7.6605e-15 0.0875 194.2857 4 2.2782e-13 0.1125 133.3333 5 3.8969e-14 0.2025 143.2099 6 7.6050e-15 0.1252 231.7104 7 -1.1369e-13 0.3250 36.9231 8 1.2612e-13 0.1350 222.2222 Iterations = 128000000 NullTime (usec) = 0.0050 MFLOPS(1) = 90.4348 MFLOPS(2) = 77.8427 MFLOPS(3) = 131.2175 MFLOPS(4) = 197.7589 AflopsWos on Rapture: 7.Ram Disk:> AflopsWOS Amiga Flops Benchmark (Double Precision), V2.0 Module Error RunTime MFLOPS (usec) 1 -7.6739e-13 0.1022 137.0031 2 -5.7021e-13 0.0823 85.0902 3 -2.4314e-14 0.0578 294.0541 4 6.8612e-14 0.0823 182.3362 5 -1.6209e-14 0.1600 181.2500 6 1.3961e-13 0.0889 326.1863 7 -3.6152e-11 0.2823 42.5131 8 8.9373e-15 0.0956 313.7255 Iterations = 256000000 NullTime (usec) = 0.0045 MFLOPS(1) = 110.8410 MFLOPS(2) = 93.8136 MFLOPS(3) = 167.9971 MFLOPS(4) = 280.3369 Last edited by grelbfarlk; 22 June 2018 at 03:22. |
13 January 2018, 22:47 | #1543 |
Posts: n/a
|
Where did you get PMC to PCI adapters?
|
13 January 2018, 23:13 | #1544 |
Registered User
Join Date: Sep 2015
Location: London, UK
Posts: 414
|
@grelbfarlk:
For a bit of a laugh, I ran flops on a 604e 166MHz (well, it seemed fast at the time ). Here's the results: FLOPS C Program (Double Precision), V2.0 18 Dec 1992 Module Error RunTime MFLOPS (usec) 1 1.3358e-12 0.2882 48.5798 2 2.0517e-13 0.2281 30.6828 3 1.7653e-14 0.1681 101.1273 4 -5.4401e-14 0.2522 59.4868 5 3.3307e-16 0.4623 62.7316 6 -1.9096e-14 0.2702 107.3386 7 2.6034e-11 0.7685 15.6153 8 -5.4068e-14 0.2882 104.1019 Iterations = 64000000 NullTime (usec) = 0.0001 MFLOPS(1) = 39.7308 MFLOPS(2) = 33.6669 MFLOPS(3) = 58.4570 MFLOPS(4) = 92.9886 |
14 January 2018, 00:02 | #1545 |
Semi-Retired
Join Date: Mar 2012
Location: Leiden / The Netherlands
Posts: 2,016
|
How does the Altivec flag work anyway? You can always compile with -S and see if there is Altivec code in it.
|
14 January 2018, 00:05 | #1546 | |
Registered User
Join Date: Dec 2015
Location: USA
Posts: 2,933
|
Quote:
https://gcc.gnu.org/onlinedocs/gcc/R...C-Options.html Without Altivec code in the source it seems to not do anything. Should probably move this to a coders thread. |
|
14 January 2018, 02:41 | #1547 |
Registered User
Join Date: Dec 2015
Location: Germany
Posts: 163
|
http://www.dyneng.com/pci2pmc.html
is this where the magic starts? Last edited by Cego; 14 January 2018 at 02:54. |
14 January 2018, 03:15 | #1548 |
Registered User
Join Date: Aug 2007
Location: USA
Posts: 362
|
|
14 January 2018, 03:20 | #1549 |
Registered User
Join Date: Dec 2015
Location: USA
Posts: 2,933
|
|
14 January 2018, 03:26 | #1550 |
Registered User
Join Date: Aug 2017
Location: USA
Posts: 728
|
lol nothing exposed thats not the name of it
|
14 January 2018, 09:53 | #1551 |
Semi-Retired
Join Date: Mar 2012
Location: Leiden / The Netherlands
Posts: 2,016
|
|
14 January 2018, 10:05 | #1552 |
Registered User
Join Date: Jun 2016
Location: Stoke-On-Trent, England
Posts: 450
|
Don't worry nothing is exposed, spent past two days on eBay looking for the thing!
No luck so far... |
14 January 2018, 10:33 | #1553 |
Registered User
Join Date: Aug 2017
Location: USA
Posts: 728
|
silkscreens are usually internal code naming eg Channel Z
|
14 January 2018, 11:37 | #1554 |
Semi-Retired
Join Date: Mar 2012
Location: Leiden / The Netherlands
Posts: 2,016
|
Last edited by Hedeon; 14 January 2018 at 11:46. |
14 January 2018, 11:54 | #1555 |
Registered User
Join Date: Aug 2007
Location: USA
Posts: 362
|
That's Awesome but isn't price and availability still a problem like with the Sonnet card?
Last edited by Magic; 14 January 2018 at 12:05. |
14 January 2018, 12:10 | #1556 |
Posts: n/a
|
prices above sonnet when you get PMC + PMC PCI carrier...
|
14 January 2018, 12:39 | #1557 |
Registered User
Join Date: Dec 2015
Location: Germany
Posts: 163
|
maybe somebody’s developing his own PMC PCI carrier? )
|
14 January 2018, 12:42 | #1558 |
Semi-Retired
Join Date: Mar 2012
Location: Leiden / The Netherlands
Posts: 2,016
|
Like I said earlier. Got my carriers and cards for 50 each (So 100 in total per combo).
|
14 January 2018, 12:56 | #1559 |
Registered User
Join Date: Aug 2017
Location: USA
Posts: 728
|
hope everyone has money left over to buy the final version which doesnt need carriers just FYI this project wasnt cheap as in time and money spent researching deving programming etc
|
14 January 2018, 13:03 | #1560 |
Registered User
Join Date: Aug 2017
Location: USA
Posts: 728
|
and if you do buy a pmc carrrier make sure it has a fan or open area for one
|
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
Thread Tools | |
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
WANTED: PCI Sound Card VORTEX2 SuperQuard / SQ2500 or Turtlebeach SantaCruz pci card | keropi | MarketPlace | 12 | 15 March 2017 02:48 |
Wanted:Sonnet Technologies QuadDoubler 50 (Doubler 4000) | kronuz | MarketPlace | 0 | 16 June 2013 00:29 |
eBay : Sonnet Quaddoubler | alexh | MarketPlace | 3 | 07 May 2013 00:50 |
Overlay problem on GeForce 7200 | Leandro Jardim | support.WinUAE | 1 | 10 August 2010 15:50 |
Prometheus PCI & Voodoo 3 PCI GFX Card | Slayer | support.Hardware | 21 | 05 September 2006 10:57 |
|
|